Knowledge Base
Understanding Cost
In-House Vs Outsource
Cost Estimation
Voice Recognition
Voice Recognition

Voice Recognition and Medical Transcription Services

Free Olympus recorder with sign up for TransDyne medical transcription services

Voice recognition services offered VS TransDyne's medical transcription services.

There has been a lot of talk about the possibility of voice recognition replacing medical transcription or at least reducing the cost incurred by medical transcription. Voice recognition technology has evolved and achieved dramatic improvisations over the years, but we have yet to see serious cost savings or benefits as a result.

This page is not a review of the technology capabilities of voice recognition; it's more a review of what medical transcription vendors who are featuring voice recognition based medical transcription services have to offer.

This review is based on a proposal made by one of the major voice recognition hardware and medical transcription service vendors (Lets Call them VRTSO) to a real hospital.

Lower the costs of your medical transcriptions with TransDyne today!

Please note that we do not have access to the actual rates quoted.


VRTSO offers a dictation platform so that doctors are able to call in and record their dictations.


There is a per-minute charge associated with each dictation made, a charge that may be many times the amount long distance charges and other telephone fees incurred by the VRTSO. This cost will also include the expense of owning, operating and supporting the dictation platform. While this cost could be in the range of 3 to 10 cents per minute, it is likely to be more based on how the VRTSO is processing the dictations.

Cost Assessment:

If VRTSO is simply handing back the dictation voice files back to the client, this cost could be as low as 3 cents per minute, on the other hand if the VRTSO is running these dictations through their backend VR engine and handing over the raw transcribed files to the client, this charge could be much more.


Dictations are run through the VRTSO's voice recognition engine and raw transcripts are handed to the client for further editing.


In addition to paying a premium for generating these raw transcripts, the client (in this case the hospital) has to use additional resources - either an in-house staff or a contracted medical transcriptions service - to edit the raw transcripts and generate functional reports of acceptable quality. This is because the quality of the raw transcripts produced is unpredictable, as is the cost of editing the raw transcripts.

Feature Considerations ∧ Expenses Relating to This Option:

1) Client will need to have tools that will enable access to the voice files as well as the ability to edit dictation from those voice files and raw transcripts. VRTSO sells or leases these tools for a set fee.

2) The VRTSO's pitch is that voice recognition efficiency improves with time, however in order to achieve a reasonable accuracy level, the client’s users (physicians) will need to adapt to and learn to dictate in a new style that the VR can process efficiently. While this is usually referred to as training the VR system, the reality is that the user must adapt to the VR systems needs.

3) Training users to dictate into the voice recognition system poses another major issue for large facilities like hospitals because there are many users who will only be using the system periodically. Each physician associated with the hospital will need to train and maintain a profile on the VRTSO’s system, and the process will need to be repeated as new residents are brought into the hospital system every year.

4) When dictating into a voice recognition system, physicians are no longer able to dictate freeform. Because raw dictation reports are produced, every little bit of formatting, as well as details of grammar and punctuation must be thought out and dictated. Similarly, if an error is made in the middle of the dictation, corrections will be extremely difficult to make.

5) It is usually assumed that for every minute of free form dictation, 3 minutes are necessary to dictate into a voice recognition system to be able to achieve efficient and accurate results. This increases the amount of time spent recording dictation, and may lower the productivity of the physicians.

TransDyne medical transcription services: Discover the convenience with a 7-day free trial.


Dictations are run through the VRTSO's VR engine and raw transcripts are edited by the VRTSO Medical Transcription Labor.


10 to 20 cents per line. It is hearsay that this particular vendor charges about 13 cents per line. Note that VRTSO charges this per line edited.

Feature Considerations ∧ Expenses Relating to This Option:

1)What constitutes a line?

a. The vendor proposal that we are referencing defines a line as a "Gross line with a minimum of 4 characters and a maximum of 65 characters which includes text in headers and footers, text in templates, text in tokens and spaces. . ." The list goes on, and the calculations lead to an increased price charged by the vendor.  Using this method of calculation, a 200,000 line document could easily become a 300,000 - 400,000 lines with a substantially higher cost as well.

2) How are edited lines counted?

a. The answer depends on how the voice recognition engine generates raw transcripts. VRTSO proposes a range X% to Y% accuracy in raw transcripts, however when those standards are not met, the responsibility falls on the user and not the system.

Let's assume that out VRTSO claimed that the raw transcripts would be 60% accurate, what does this mean to the client? Will 40% of each document need to be edited? Will 4 of every 10 pages need to be edited completely?

If 40% of each line of raw transcript needs to be edited (which would mean that each and every page would need to be edited), the client would have to pay for the entire line count, however inaccurate of edits.

On the other hand, if only 40% of the raw transcripts need to be edited, the client would be responsible for the cost of editing 120,000 - 160,000 lines.

Discover the value of low cost medical transcriptions from TransDyne!


Monthly rental equipment: Rather than paying for the VRTSO to edit raw dictation files, the client will be provided equipment for managing work flow and will use let internal resources to edit the raw files.


VRTSO in this case, charges the client pay for every minute of usage while they call in their dictation and then goes on to add a monthly rental charge for each piece of equipment that they lease. An estimation is that it costs $800 per month for 10 sets of equipment and licenses.

Feature Considerations ∧ Expenses Relating to This Option:

Additional maintenance costs - as well as potential equipment downtime and employee time off can create an even greater expense for the client.

In summary, this VRTSO uses an extremely expensive line count methodology, cannot guarantee raw transcript quality and the service adds a burden on physicians working for the client; in this case, the hospital.

Rather than making an effort to lower the costs of medical transcription, this hospital would inevitably increase their costs by working with the VRTSO.

Try us free for 7 days!


      © TransDyne 2009